Choose Rich Live
Tech

Google launches Gemini Decisions API to evaluate closed-set choices without generating text

Google has launched the Gemini Decisions API, a service powered by its Gemini Decision 1 model that scores up to 1,000 closed-set questions across multimodal inputs in a single pass with zero output token fees.

Diagram explaining Google's Gemini Decisions API workflow, showing multimodal inputs and questions feeding into Gemini Decision 1 to generate calibrated answers.
Image: @Cointelegraph

Google has introduced the Gemini Decisions API, powered by a new specialized model called Gemini Decision 1, according to company announcements and reports by Techmeme and The New Stack. The API trades conversational text generation for rapid, structured outputs, returning predefined answers such as yes-or-no responses, multiple-choice selections, and numerical ratings accompanied by confidence scores.[1][3][4][6][8]

The model features a 256K-token context window and handles multimodal inputs including text, JSON, PDFs, images, audio, and video, according to diagrams shared by Cointelegraph and posts by Ashutosh Shrivastava. In a single pass, the API can evaluate up to 1,000 closed-set questions using parallel isolated scoring, which Google expects will streamline decision-making in policy enforcement and AI-driven workflows.[1][2][4][6][7]

Google is pricing the service at 10 cents per million input tokens with no charge for output tokens, Techstrong.ai reported. Google reported median compute latencies between 50 and 100 milliseconds and claimed the API is nearly five times cheaper than its entry-level Gemini 3.5 Flash-Lite, positioning Google directly against rivals in the decision-model category including TypeSafe and OpenAI.[3][5][7][8]

Key facts

  • Google launched the Gemini Decisions API, powered by a model called Gemini Decision 1.
  • The API returns yes-or-no responses, multiple-choice selections, and numerical ratings with confidence scores instead of generated text.
  • The service can evaluate up to 1,000 closed-set questions in a single pass.
  • Gemini Decision 1 features a 256K-token context window and supports text, JSON, PDFs, images, audio, and video inputs.
  • The API is priced at $0.10 per million input tokens with no charge for output tokens.
  • Google claims median compute latency is 50 to 100 milliseconds and says the service is nearly five times cheaper than Gemini 3.5 Flash-Lite.
  • The launch places Google in competition with decision-oriented models from TypeSafe and OpenAI.

Sources · 8 sources

  1. TN

    The New Stack@thenewstackPost on X ·

    Google’s new Gemini Decisions API makes fast, low-cost decisions without generating text, scoring up to 1,000 closed-set choices in a single pass. https://t.co/Kki3CJbCjN

    Open source
  2. CB

    Crypto BriefingArticle ·

    Google launches Gemini Decisions API for fast multimodal choices Google's Gemini Decisions API could streamline decision-making processes, enhancing efficiency in AI-driven applications and policy enforcement. The post Google launches Gemini Decisions API for fast multimodal choices appeared first on Crypto Briefing .

    Open source
  3. TA

    Techstrong.ai@TechstrongaiPost on X ·

    Google has launched the Gemini Decisions API, a specialized service powered by Gemini Decision 1 that trades the broad conversational abilities of traditional LLMs for rapid, structured outputs at a fraction of the price, returning yes-or-no answers, multiple-choice selections and numerical ratings with confidence scores instead of prose. At 10 cents per million input tokens with no charge for output, Google claims it is nearly five times cheaper than its entry-level Gemini 3.5 Flash-Lite, with median compute latency of 50 to 100 milliseconds, placing Google directly into a decision-model market where TypeSafe's Jev and OpenAI are already competing. Here's the full breakdown of the benchmarks and pricing: https://t.co/7G3vhcVOuP #AI #Google #Gemini #EnterpriseAI #LLM

    Open source
  4. TE

    TechmemeArticle ·

    Google launches Gemini Decisions API, powered by its new Gemini Decision 1 model, to return predefined answers with confidence scores, at $0.10/1M input tokens (Google) Google : Google launches Gemini Decisions API, powered by its new Gemini Decision 1 model, to return predefined answers with confidence scores, at $0.10/1M input tokens — Introducing the Gemini Decisions API, powered by our new Gemini Decision 1 model. Evaluate up to 1,000 yes/no, choice …

    Open source
  5. SI

    SiliconANGLE@SiliconANGLEPost on X ·

    Google enters the decision model market with Gemini Decisions API https://t.co/xkKc7AZ4cK

    Open source
  6. CO

    Cointelegraph@CointelegraphPost on X ·

    🤖 NEW: Google has launched its Gemini Decisions API, allowing AI to evaluate multimodal inputs against up to 1,000 questions in a single pass with zero output tokens billed. https://t.co/Wbq7ZynccX

    Open source
  7. AS

    AshutoshShrivastava@ai_for_successPost on X ·

    Google just dropped its first decision-making model Gemini Decision 1 in the Gemini API! - 256K-token context window - Supports images, PDFs, audio, and video - $0.10 per 1M input tokens with zero output tokens billed. One example of how Google's team is using it: Stitch Rapid uses the Decisions API to design an entire product flow in seconds (1x speed). https://t.co/eXsRdizRTF

    Open source
  8. MT

    MTS@MTSlivePost on X ·

    SITUATION DETECTED: Google announced the Gemini Decisions API, powered by their new model Gemini Decision 1. https://t.co/xsp0cOTcDw

    Open source